Cancer is the most commonly occurring and deadly genetically-based disease. It is increasingly being recognized that the genetic changes that bring about the transformation and progression of a normal cell to cancer, are imprinted on the genome architecture. By comprehensive characterization of the cancer genome, we can read and interpret the recorded changes with respect to the mechanisms that bring about the complete reprogramming of the normal cell into a fatal disease. Recent work has focused attention on our expanding abilities to interrogate the genome properties at ever increasing resolution, until now we arrive at the base pair level. The BCM Tumor Genome Characterization Center aims to analyze sets of tumors plus, when appropriate, matched normal tissue to characterize and enumerate the somatic changes occurring in 500 patients for each of 20-25 tumor types over the next 5 years. The characterization will be accomplished by DNA sequencing alone using the AB/SOLiD platform already in scaled-up production in the Baylor College of Medicine Human Genome Sequencing Center, the performance site for this application. The characterizations will include genome-wide expression levels with analysis of individual mRNA molecules to look for recurrent aberrant splicing and gene fusion, and eventual analysis of bi-allellic expression for heterozygous loci. We will also perform copy number variation in genomic DNA to look for recurrent amplification or deletion of genomics segments with resolution down to 10 kb, with inclusion of diTag libraries by year 3 of the grant that will enable precise breakpoint detection of insertions deletions inversions and translocations. There are two principal advantages to our proposed sequencing approach. First, it will avoid a complex transition of the program from a chip platform to a sequencing platform in the early years of the program. The complexity of transition would include expensive dismantling and reconstructing validated laboratory and informatic pipelines, and possible disruption of service. Second, sequencing data will be immediately comparable and complimentary with whole genome sequencing approaches that will be conducted in parallel by the NHGRI, in which the BCM HGSC will be a participant.